Search

Much of political science involves the study of causal mechanisms, and causal mediation analysis has grown rapidly across different subfields over the past decade. Yet, conventional methods for analyzing causal mechanisms are difficult to use when the causal effect of interest involves multiple mediators that are potentially causally dependent—a common scenario in political science applications. This article introduces a general framework for tracing causal paths with multiple mediators. In this framework, the total effect of a treatment on an outcome is decomposed into a set of path-specific effects (PSEs). We propose an imputation approach for estimating these PSEs from experimental and observational data, along with a set of bias formulas for conducting sensitivity analysis. We illustrate this approach using an experimental study on issue framing effects and an observational study on the legacy of political violence. An open-source R package, paths, is available for implementing the proposed methods.

A critical barrier to generating cumulative knowledge in political science and related disciplines is the inability of researchers to observe the results from the full set of research designs that scholars have conceptualized, implemented, and analyzed. For a variety of reasons, studies that produce null findings are especially likely to be unobserved, creating biases in publicly accessible research. While several approaches have been suggested to overcome this problem, none have yet proven adequate. We call for the establishment of a new discipline-wide norm in which scholars post short "null results reports" online that summarize their research designs, findings, and interpretations. To address the inevitable incentive problems that earlier proposals for reform were unable to overcome, we argue that decentralized research communities can spur the broader disciplinary norm change that would bring advantage to scientific advance. To facilitate our contribution, we offer a template for these reports that incorporates evaluation of the possible explanations for the null findings, including statistical power, measurement strategy, implementation issues, spillover/contamination, and flaws in theoretical priors. We illustrate the template's utility with two experimental studies focused on the naturalization of immigrants in the United States and attitudes toward Syrian refugees in Jordan., Political Analysis, 31 (3), ISSN:1047-1987, ISSN:1476-4989

The COVID-19 pandemic imposed new constraints on empirical research, and online data collection by social scientists increased. Generalizing from experiments conducted during this period of persistent crisis may be challenging due to changes in how participants respond to treatments or the composition of online samples. We investigate the generalizability of COVID era survey experiments with 33 replications of 12 pre-pandemic designs, fielded across 13 quota samples of Americans between March and July 2020. We find strong evidence that pre-pandemic experiments replicate in terms of sign and significance, but at somewhat reduced magnitudes. Indirect evidence suggests an increased share of inattentive subjects on online platforms during this period, which may have contributed to smaller estimated treatment effects. Overall, we conclude that the pandemic does not pose a fundamental threat to the generalizability of online experiments to other time periods.

With the availability of large volumes of electronic communications data and the increasing sophistication of computational techniques, the development of automated approaches for different kinds of framing analysis is an important goal of researchers. There is as yet no standard method for the ‘unsupervised’ inductive identification of frames based upon the content of articles. Three groups of core approaches underlie a wide range of work in this area, and we compare three techniques based on these approaches against each other and against manual human analysts. The three techniques are a k-means clustering algorithm together with a sophisticated natural language processing (NLP)-based feature selection process; evolutionary factor analysis (EFA), a factor analysis approach; and the structural topic model (STM). We use two datasets – one very focused and one extremely broad – as examples of the kind of frame analysis problems readers may wish to attempt. Even in a highly targeted dataset, we find some distance between the frames generated by computational analysis and those manually produced by our human analysts. The details of each method have a substantial impact on frame quality and interpretation. We find that the STM approach is the most effective for our narrow-scope dataset, but that it returns definite topics, and not frames, when working on our very broad dataset. We also show that we can get results emphasizing different parts of the framing problem by combining parts of different methods as a multi-stage process, rather than viewing available methods as simple plug-and-play models.

The April 2018 article of Diana Mutz “Status Threat, Not Economic Hardship, Explains the 2016 Presidential Vote,” was published in the Proceedings of the National Academy of Sciences and contradicts prior sociological research on the 2016 election. Mutz’s article received widespread media coverage because of the strength of its primary conclusion, declaimed in its title. The present article is a critical reanalysis of the models offered by Mutz, using the data files released along with her article. Contrary to her conclusions, this article demonstrates that (1) the relative importance of economic interests and status threat cannot be estimated effectively with her cross-sectional data, and (2) her panel data are consistent with the claim that economic interests are at least as important as status threat. The preexisting sociological literature has offered interpretations that incorporate economic interests and, as a result, provides a more credible explanation of the 2016 election.

We develop a new measure of party position based on a scaling of ideology tags supplied in infoboxes on political parties’ Wikipedia pages. Assuming a simple model of tag assignment, we estimate the locations of parties and ideologies in a common space. We find that the recovered scale can be interpreted in familiar terms of “left versus right.” Estimated party positions correlate well with ratings of parties’ positions from extant large-scale expert surveys, most strongly with ratings of general left–right ideology. Party position estimates also show high stability in a test–retest scenario. Our results demonstrate that a Wikipedia-based approach yields valid and reliable left–right scores comparable to scores obtained via conventional expert coding methods. It thus provides a measure with potentially unlimited party coverage. Our measurement strategy is also applicable beyond Wikipedia.

Voters are frequently ill-equipped to monitor politicians’ actions. Politicians are expected to implement projects, whose benefits sometimes partially accrue to interest groups (IGs) and not entirely to voters. IGs thus have an incentive to affect which projects politicians implement by providing implementation subsidies to lower the cost of policymaking that politicians incur. This article shows how these considerations interact in a two-period political agency model with moral hazard and adverse selection. I study how the involvement of IGs in the policymaking process can affect voters’ welfare and show why voters might rationally not perfectly monitor politicians in the presence of IGs that might capture projects’ benefits and affect policymaking. (JEL D72, D73, D83)

Citizens are increasingly involved in the design of democratic institutions, for instance via referendums. If they support the institution that best serves their self-interest, the outcome inevitably advantages the largest group and disadvantages minorities. In this paper, we challenge this pessimistic view with an original lab experiment in France and Great Britain. In the first phase, experimental subjects experience elections under plurality and approval voting. In the second phase, they decide which rule they want to use for extra elections. The treatment is whether they do or do not have information to determine where their self-interest lies before deciding. We find that self-interest shapes people’s decisions, but so do intrinsic egalitarian values that subjects have outside of the lab. The implications are: (1) people have consistent ‘value-driven preferences’ for electoral rules, and (2) putting them in a situation of uncertainty leads to an outcome that reflects these values.

The prespecification of the network is one of the biggest hurdles for applied researchers in undertaking spatial analysis. In this letter, we demonstrate two results. First, we derive bounds for the bias in nonspatial models with omitted spatially-lagged predictors or outcomes. These bias expressions can be obtained without prior knowledge of the network, and are more informative than familiar omitted variable bias formulas. Second, we derive bounds for the bias in spatial econometric models with nondifferential error in the specification of the weights matrix. Under these conditions, we demonstrate that an omitted spatial input is the limit condition of including a misspecificed spatial weights matrix. Simulated experiments further demonstrate that spatial models with a misspecified weights matrix weakly dominate nonspatial models. Our results imply that, where cross-sectional dependence is presumed, researchers should pursue spatial analysis even with limited information on network ties.

Using a natural experiment, we find that in provinces where Turkey’s Islamic Justice and Development Party (AKP) just won the election in 2004, women, including those who are weakly religious or non-religious, now veil far more than in provinces in which AKP just lost, the more so the poorer they are. This effect, as we predict, does not occur for praying regularly which is more costly and harder to observe practice. We argue that veiling is higher in AKP provinces not only because of a generic aim to conform to the stricter mores fostered by the victorious party. We find that those who veil, particularly those in AKP provinces who are not pious, are more politically active than those who do not veil. This may be an indication that veiling could partly be a strategic response to policies, which favour those who are or appear pious. Our study suggests that observable religious practices may have their independent dynamics driven by the pursuit of instrumental goals. Our results also suggest that parties with a religious ideology have an advantage over their secular counterparts in solving the clientelistic information problem, for they can rely on religious symbols for screening and signalling.

Many academic surveys administered online include a banner along the top of the survey displaying the name or logo of the researcher’s university. These banners may unintentionally influence respondents’ answers since subtle contextual cues often have great impact on survey responses. Our study aims to determine whether these banners influence survey respondents’ answers, that is, whether they induce sponsorship effects. For this purpose, we field three different studies on Amazon’s MTurk where we randomly assign the sponsoring institution. Our outcome measures include survey questions about social conservatism, religious practices, group affect, and political knowledge. We find that respondents provide similar answers and exhibit similar levels of effort regardless of the apparent sponsor.

In the recent decades, privacy scholarship has made significant progress. Most of it was achieved in monodisciplinary works. However, privacy has a deeply interdisciplinary nature. Most importantly, societies as well as individuals experience privacy as being influenced by legal, technical, and social norms and structures. In this article, we hence attempt to connect insights of different academic disciplines into a joint model, an Interdisciplinary Privacy and Communication Model. The model differentiates four different elements: communication context, protection needs, threat and risk analysis, as well as protection enforcement. On the one hand, with this model, we aim to describe how privacy unfolds. On the other hand, the model also prescribes how privacy can be furnished and regulated. As such, the model contributes to a general understanding of privacy as a theoretical guide and offers a practical basis to address new challenges of the digital age.

This article offers a rationale for candidates who voluntarily pledge to term limits. My analysis is built on a standard, two-period political agency model to which I add an election campaign where candidates can commit not to seek a second term. Pledging to term limits allows candidates to signal their private type and insulate themselves from career concerns. By doing so, candidates leverage the fact that the representative voter prefers to elect a candidate who does not seek reelection because she either has more desirable characteristics, distorts her decisions to a lesser extent, or both. As a result, candidates who pledge to term limits have a higher probability of being elected in the first place. I characterize the equilibria of a model specification in which politicians differ with respect to their policy preferences and uncover circumstances in which term limits pledges are informative and can improve the representative voters welfare.

Across the social sciences, generalized trust has been one of the most frequently examined constructs since researchers first introduced measures of it in the 1950s. Despite its significance, there is growing consensus that conventional measures of generalized trust are prone to measurement invalidity and non-equivalence, which places sociological knowledge about generalized trust in serious doubt. In this article, I advance trust research in sociology by (1) refining two new self-report measures of generalized trust—the Stranger Face Trust scale (SFT) and the Imaginary Stranger Trust scale (IST)—and (2) assessing their empirical performance on a nationally representative probability sample (N = 1,264). I compare the reliability and validity of SFT, IST, and traditional measures of generalized trust across a number of measurement validation tests (convergent validity, discriminant validity, known-groups validity, and predictive validity). Results suggest that SFT provides the most accurate and consistent measure of generalized trust. I conclude by discussing the implications of my findings for sociological knowledge.

Political scientists have long been interested in the persistent effects of history on contemporary behavior and attitudes. To estimate legacy effects, studies often compare people living in places that were historically exposed to some event and those that were not. Using principal stratification, we provide a formal framework to analyze how migration limits our ability to learn about the persistent effects of history from observed differences between historically exposed and unexposed places. We state the necessary assumptions about movement behavior to causally identify legacy effects. We highlight that these assumptions are strong; therefore, we recommend that legacy studies circumvent bias by collecting data on people's place of residence at the exposure time. Reexamining a study on the persistent effects of US civil-rights protests, we show that observed attitudinal differences between residents and non-residents of historic protest sites are more likely due to migration rather than attitudinal change.

Research in political science has shown that UN peacekeeping operations are an important tool for ending civil war violence. However, much less is known about how UN peacekeepers affect civilian victimization. Given that civilians bear the primary costs of intrastate conflict, understanding how international actors can contribute to the resolution of violence affecting them is a pressing concern. How does the presence of UN peacekeepers affect civilian victimization? We address this question by offering a straightforward empirical test of how UN peacekeeping patrols affect the likelihood that there will be violence against civilians. We build on the existing literature and established practices of peacekeeping to argue that peacekeepers deter violence against violence. To test our argument, we examine the case of Mail, the site of large-scale communal violence managed by UN peacekeepers since 2013. We employ a Geographic Regression Discontinuity Design (GRDD) around the border of Mali and Burkina Faso to estimate the causal effect of deploying peacekeepers to an area with growing communal tensions. Ultimately, our research provides robust causal evidence that UN peacekeeping works at the local level to protect civilians.

This review systematizes the emerging literature for causal inference using deep neural networks under the potential outcomes framework. It provides an intuitive introduction on how deep learning can be used to estimate/predict heterogeneous treatment effects and extend causal inference to settings where confounding is non-linear, time varying, or encoded in text, networks, and images. To maximize accessibility, we also introduce prerequisite concepts from causal inference and deep learning. The survey differs from other treatments of deep learning and causal inference in its sharp focus on observational causal estimation, its extended exposition of key algorithms, and its detailed tutorials for implementing, training, and selecting among deep estimators in Tensorflow 2 available at github.com/kochbj/Deep-Learning-for-Causal-Inference.

This paper assesses the performance of regularized generalized least squares (RGLS) and reweighted least squares (RLS) methodologies in a confirmatory factor analysis model. Normal theory maximum likelihood (ML) and GLS statistics are based on large sample statistical theory. However, violation of asymptotic sample size is ubiquitous in real applications of structural equation modeling (SEM), and ML and GLS goodness-of-fit tests in SEM often make incorrect decisions on the true model. The novel methods RGLS and RLS aim to correct the over-rejection by ML and under-rejection by GLS. Proposed by Arruda and Bentler (2017), RGLS replaces a GLS weight matrix with a regularized one. Rediscovered by Hayakawa (2019), RLS replaces this weight matrix with one that derives from an ML function. Both of these methods outperform ML and GLS when samples are small, yet no studies have compared their relative performance. A confirmatory factor analysis Monte Carlo simulation study with normal and non-normal data was carried out to examine the statistical performance of these two methods at different sample sizes. Based on empirical rejection frequencies and empirical distributions of test statistics, we find that RLS and RGLS have equivalent performance when N≥70; whereas when N

Ex post review is a common feature of policymaking institutions. We consider an environment in which an expert agent makes a policy recommendation, which can then be accepted or rejected by an overseer whose policy goals differ from those of the agent. The theory suggests that both behavior and optimal institutional design are sensitive to several factors, including actors' preference alignment, the importance of the policy decision, and the uncertainty about the correct policy choice. We characterize the types of situations in which ex post review creates incentives for the agent to make pathological policy choices. In these situations, ex post review can reduce accountability of the agent to overseer wishes and ultimately provide incentives to set aside review entirely. The theory also offers testable predictions about policy recommendations and the overseer's acceptance or rejection of these recommendations.

Scientific knowledge on the effectiveness of governmental military build-ups to terminate intrastate conflicts is sparse and inconclusive. Developing a war-of-attrition framework, we derive the impact of governments' armaments on the duration of these conflicts: military build-ups, as reflected in inflows of major conventional weapons, enable the government to inflict costs onto its adversaries, forcing them to withdraw earlier from the conflict. This type of weapons is required in particular to project military power over larger distances and to fight rebels in remote areas. Using SIPRI arms transfer data for the first time in a disaggregated dyadic design, covering 418 government-rebel group dyads in 134 conflicts, we empirically corroborate the formal model's predictions. As endogeneity issues might arise when governments procure arms in anticipation of a protracted conflict, we ensure causal identification with an instrumental variable survival approach based on the Aalen additive hazards model.

In this essay I argue that the need to motivate bureaucrats to invest in high quality policy implementation alters the appeal of executive unilateralism. If executive orders are less durable than legislation, then the bureaucracy will have weaker incentives to invest in policymaking. This affects presidents’ willingness to compromise and work with Congress to pass legislation, rather than pursue unilateral action. Unilateralism becomes less attractive as bureaucratic capacity increases and as the relative durability of executive orders decreases. I formalize this logic in a simple formal model and discuss fruitful extensions for future work.

We develop a theory of policymaking between an agent and an overseer, with a principal whose welfare is affected by agent-overseer interactions. The agent can increase the quality of policy outcomes through costly capacity investments. Oversight and agent bias jointly determine optimal agent capacity investments. We show that when oversight improves agent investment incentives the principal always benefits from an agent with biases opposite the overseer. Competing agent-overseer biases translate into higher quality policy outcomes than the principal could induce were she monitoring the agent. Effective oversight is necessary for these incentive effects. The results imply that political principals ought to consider the nature of the broader policymaking environment when appointing agents to make policy on their behalf and when designing managerial strategies aimed at motivating agents.

How does the scope of review affect bureaucratic policymaking? To explore this question, I consider a policymaking environment in which an expert agency develops policy that is upheld or overturned by an overseer who may have different policy goals. The agency can affect the quality of implementation through effort investments in addition to choosing the substantive content of policy. Under procedural review the overseer only reviews the agency’s effort, which allows the agency to fully utilize its expertise. Substantive review also tasks the overseer with judging agencies’ substantive policy choices, which can lead the agency to disregard its superior information and obfuscate to avoid reversal. Depending on the policy environment, this dynamic can either benefit or harm the overseer. In some cases the overseer can be made better off by having less transparent review institutions; that is, institutions that direct the overseer to only review procedure and preclude judging substance.

Does online political involvement reinforce or compensate participatory deficiencies at the polls? Extant survey evidence portrays online participation as a weapon of the strong, wielded by a highly politically involved, white, and affluent subset of the American electorate. Surveys face systematic sampling and measurement errors in the domain of political participation, however. In this study, I revisit this question using individual voter registration records that I integrate with observed Twitter activity. Based on a large sample that reflects Florida's voting-eligible population, I find that political involvement on Twitter is prevalent across the electorate and extends to those most likely to abstain from voting. Moreover, race and income, which are salient dividing lines in voting, do not structure social media-based political participation, and common turnout patterns for age and party subgroups are reversed, though especially among more engaged voters. These results offer a novel perspective on reinforcement theory and social media's compensatory potential for more inclusive representation. I discuss implications for political representation and future research examining political involvement.

Climate change concern varies widely across countries. In 2019, 80% of Greeks were at least somewhat worried about climate change, compared to 20% of Egyptians. We argue that variation in climate change concern is influenced by regime type as providing civil liberties limit government’s ability to control the public discourse on anthropogenic climate change. We provide empirical evidence for the effect of democracy on climate change concern using data from 611,909 individuals from 118 countries collected between 2007 and 2019. Exploit-ing variation across country-cohorts and time, we find one unit change in the 7-point civil liberty index to influence climate change concern by 2.3 [95% CI: ±1] percentage points. The effect is much stronger in wealthy countries and less educated cohorts. We also present evidence for our causal pathway using qualitative interviews and by modeling the association between democracy, climate protest, climate media coverage, and climate concern with simultaneous equations.

Interactions between units in political systems often occur across multiple relational contexts. These relational systems feature interdependencies that result in inferential shortcomings and poorly-fitting models when ignored. General advancements in inferential network analysis have improved our ability to understand relational systems featuring interdependence, but developments specific to working with interdependence that cross relational contexts remain sparse. In this paper, I introduce a multilayer network approach to modeling systems comprising multiple relations using the exponential random graph model. In two substantive applications, the first a policy communication network and the second a global conflict network, I demonstrate that the multilayer approach affords inferential leverage and produces models that better fit observed data.

What rhetorics run throughout radical discourse, and why do some gain prominence over others? The scholarship on radicalism largely portrays radical discourse as opposition to powerful ideas and enemies, but radicals often evince great interest in personal and local concerns. To shed light on how radicals use and adopt rhetoric, we analyze an original corpus of more than 23,000 pages produced by Afghan radical groups between 1979 and 2001 using a novel computational abductive approach. We first identify how radicalism not only attacks dominant ideas, actors, and institutions using a rhetoric of subversion, but also how it can use a rhetoric of reversion to urge intimate transformations in morals and behavior. Next, we find evidence that radicals’ networks of support affect the rhetorical mixture they espouse, due to social ties drawing radicals into encounters with backers’ social domains. Our study advances a relational understanding of radical discourse, while also showing how a combination of computational and abductive methods can help theorize and analyze discourses of contention.

Are offline political relations and endorsements established or maintained with online interactions?We examine the external validity of data on political behavior exhibited online and compare it to offline behavior. We overlay two data sets on political support among members of the Swiss parliament to estimate the explanatory power of online endorsements on offline cosponsorship signatures among members, using a gHypEG network regression model.Whereas offline support behavior is driven by a broad range of factors, such as network and homophily effects, online endorsement is less diverse.Our findings show that online endorsement is predominantly driven by partisanship, calling into question the use of online political behavior data to study real-world politics.

Computational methods have become an integral part of political science research. However, helping students to acquire these new skills is challenging because programming proficiency is necessary, and most political science students have little coding experience. This article presents pedagogical strategies to make transitioning from Excel, SPSS, or Stata to R or Python for data analytics less challenging and more exciting. First, it discusses two approaches for making computational methods accessible: showing the big picture and walking through the workflow. Second, a step-by-step guide for a typical course is provided using three examples: learning programming fundamentals, wrangling messy data, and communicating data analysis.

An interaction in a fixed effects (FE) regression is usually specified by demeaning the product term. However, this strategy does not yield a genuine within estimator. Instead, an estimator is produced that reflects unit-level differences of interacted variables whose moderators vary within units. This is desirable if the interaction of one unit-specific and one time-dependent variable is specified in FE, but it may yield problematic results if both interacted variables vary within units. Then, as algebraic transformations show, the FE interaction estimator picks up unit-specific effect heterogeneity of both variables. Accordingly, Monte Carlo experiments reveal that it is biased if one of the interacted variables is correlated with an unobserved unit-specific moderator of the other interacted variable. In light of these insights, we propose that a within interaction of two timedependent variables be estimated by first demeaning each variable and then demeaning the product term. This 'double-demeaned' estimator is not subject to bias caused by unobserved effect heterogeneity. It is, however, less efficient than standard FE and only works with T>2.

The idea that electoral competition, party promises and manifestos are important for how the representative behaviour of parties unfolds post-electorally is central to democratic theories of representation. Scholars of Party Mandate fulfilment have for long been focusing methodologically on the number of pre-electoral pledges parties in government manage to realize. This paper analyses existing research and its limitations and proposes a novel approach for mandate fulfilment, extending past research (both conceptually and methodologically) by looking at the extent to which the parliamentary discourse of parties matches their electoral discourse. It designs, tentatively validates and implements a novel measure of how similar is the pre and post-electoral rhetoric of parties by utilizing recent advances in computational linguistics and the diffusion of text analysis tools in the social sciences. It applies it on the Irish parliament and compares electoral manifesto data with post-electoral parliamentary data from 1997 to 2019.

Hand-labeled political texts are often required in empirical studies on party systems, coalition building, agenda setting, and many other areas in political science research. While hand-labeling remains the standard procedure for analyzing political texts, it can be slow and expensive, and subject to human error and disagreement. Recent studies in the field have leveraged supervised machine learning techniques to automate the labeling process of electoral programs, debate motions, and other relevant documents. We build on current approaches to label shorter texts and phrases in party manifestos using a pre-existing coding scheme developed by political scientists for classifying texts by policy domain and policy preference. Using labels and data compiled by the Manifesto Project, we make use of the state-of-the-art Bidirectional Encoder Representations from Transformers (BERT) in conjunction with Convolutional Neural Networks (CNN) and Gated Recurrent Units (GRU) to seek out the best model architecture for policy domain and policy preference classification. We find that our proposed BERT-CNN model outperforms other approaches for the task of classifying statements from English language party manifestos by major policy domain.

Recent years have seen major political crises throughout the world. Most recently, the US was swept by a wave of protests, urban riots, and violent confrontations between left- and right-wing extremists. Understanding how future crises will unfold and assessing the resilience of different countries to various shocks is of foremost importance in averting the human costs of state breakdown and civil war. In a recent publication (Turchin et al. 2018) we proposed a novel transdisciplinary approach to modeling social breakdown, recovery, and resilience. This approach builds on recent breakthroughs in macrosocial dynamics (and specifically structural-demographic theory), statistical analysis of large-scale historical data, and dynamic modelling. Our main goal is to construct a series of probabilistic scenarios of social breakdown and recovery. We called this approach—similar to ensemble forecasting in weather prediction—multipath forecasting (MPF). In this article I develop a “prototype” of the MPF engine with the goal of illustrating the utility a fully developed version may have. I first apply the computational model to the period of American history from the beginning of the nineteenth to the end of the twentieth century, with the goal of parameterizing the model and testing it against data. Then I use the parameterized model to forecast the dynamics of instability in the USA beyond 2020 and illustrate how the MPF engine can be used to explore the effects of different policy interventions.

The article discusses fascism's place on the political spectrum. At present, there is no consensus among political scientists and economists on that issue, as it has been extraordinarily politicized and distorted during ideological struggles among various currents of socialism. From the very beginning, fascism was depicted by Marxists as belonging to the Right, while Fascists themselves wanted to build a society that transcends the Left-Right paradigm. However, few voices in academia have noted that practical implementation of the fascists’ ideas, inherited from the works of revolutionary and national syndicalists, exhibited predominantly leftist characteristics.The ambiguity of placing fascism in its proper place on the political spectrum can be confidently resolved by applying three primary factors that govern political spectrum polarization: attitude to private property, scope of individual freedom, and degree of wealth redistribution. The article argues that fascism is a particular current of non-Marxian socialism that utilized collectivization of consciousness and wealth redistribution as the main paths toward socialism rather than outright expropriation of private property or means of production. Simultaneously, it is acknowledged that private property rights were inhibited by the fascist state, even though de jure they were permitted.The fascist ideal of the “alternate way” had a logical inconsistency that produced an unstable equilibrium between labor and capital as well as between the man and the state. The politico-economic structure predictably collapsed to the left in the course of building a new society. Therefore, fascism could be correctly called the Right of the Left.

Human decision-making is affected by a diversity of factors including material cost–benefit considerations, normative and cultural influences, learning and conformity with peers and external authorities (e.g. cultural, religious, political, organisational). Also important are dynamically changing personal perceptions of the situation and beliefs about actions and expectations of others as well as psychological phenomena such as cognitive dissonance and social projection. To better understand these processes, I develop a unifying modelling framework describing the joint dynamics of actions and attitudes of individuals and their beliefs about the actions and attitudes of their groupmates. I consider which norms get internalised and which factors control beliefs about others. I predict that the long-term average characteristics of groups are largely determined by a balance between material payoffs and the values promoted by the external authority. Variation around these averages largely reflects variation in individual costs and benefits mediated by individual psychological characteristics. The efforts of an external authority to change the group behaviour in a certain direction can, counter-intuitively, have an opposite effect on individual behaviour. I consider how various factors can affect differences between groups and societies in the tightness/looseness of their social norms. I show that the most important factors are social heterogeneity, societal threat, effects of authority, cultural variation in the degree of collectivism/individualism, the population size and the subsistence style. My results can be useful for achieving a better understanding of human social behaviour and historical and current social processes, and in developing more efficient policies aiming to modify social behaviour.

Track two diplomacy is the set of government-sanctioned but informal diplomatic activities that are intended to overcome difficulties associated with formal diplomacy. Track two meetings should complement official diplomacy because they are external to the government apparatus and have low public visibility, insulating governments from unwanted commitment and attention. While track two meetings are a prevalent diplomatic tool, empirical examinations of their effectiveness using larger-scale statistical approaches are rare. Using a multilayer network approach, I explore the relationship between official and track two diplomacy in the Asia Pacific region using a new network data set constructed from diplomatic meeting records. Results indicate that the two tracks of diplomacy are mutually reinforcing, which suggests that track two diplomacy is effective in facilitating formal international cooperation. Further examination finds evidence to support the notion that states strategically select between different tracks of diplomacy to avoid being publicly tied to unpalatable opponents.

In recent decades, another metric has come to the fore that takes a fundamentally different approach to evaluation. This approach measures impact, inferred from the citations accrued by a study. Like reputation, impact may be measured at any level. An individual may be evaluated according to the citations accruing to her publications; a department may be evaluated according to the citations accrued by studies produced by all its faculty; a journal or press may be evaluated according to the citations accrued by all its publications; and so forth. Naturally, there is some inter-dependence among impact measures. A study’s citation-count is apt to be influenced by a journal’s citation-count, for example. Nonetheless, there is a much greater degree of independence among the units being assessed. An implication is that impact-based metrics are more meritocratic than reputational metrics

How do politicians gain an upper hand in political discourse? Recent literature on framings in political discourse have re-directed attention from static models of issue ownership to more interactive models of strategic framing contests. This article proposes a robust action approach, and tests hypotheses about the dual repertoires of local action: First, we test for the presence of exploratory local action, when discourse positions are altered in response to positions by others, and second, we show evidence for the presence of role claim behavior, when a dominant discourse position is taken that silences others. Using the case of economic policy discourse in Hungary in 1997, we show how the “GDP growth” discourse position silenced opposition positions on the right, mostly built from stigmatizing frames. Discourse positions beyond the one built around growth did not silence alternative framings, but elicited discourse shifts. We coded 8632 utterances over 100 days of discourse into a two-mode network of speech acts and statements and used a two-mode blockmodeling approach to identify positions and frames. We used a pooled time series approach to test hypotheses of local action dynamics. We found evidence for both exploratory local action and role claim, while controlling for observed and unobserved heterogeneity at the actor level.

This study examines how people with disabilities and chronic health conditions—members of a large and diverse group often overlooked by Canadian public policy—are making sense of the Canadian federal government’s response to COVID-19. Using original national online survey data collected in June 2020 (N= 1,027), we investigate how members of this group view the government’s overall response. Although survey results show broad support for the federal government’s pandemic response, findings also indicate fractures based on disability type and specific health condition, political partisanship, region, and experiences with COVID-19. Among these, identification with the Liberal party and receipt of CERB stand out as associated with more positive views. Further examination of qualitative responses shows that these views are also linked to differing perspectives surrounding government benefits and spending, partisan divisions, and other social and cultural cleavages. La présente étude porte sur la façon dont les personnes atteintes de handicap ou de maladie chronique — appartenant à un vaste groupe diversifié, souvent ignoré des politiques publiques canadiennes — interprètent la réaction du gouvernement fédéral canadien à la COVID‑19. À l’aide de données nationales exclusives tirées d’un sondage en ligne effectué en juin 2020 (N= 1 027), nous analysons le point de vue des membres de ce groupe quant à la réaction globale du gouvernement. Bien que les résultats du sondage témoignent d’un large appui à la réaction du gouvernement fédéral à la pandémie, nos observations révèlent également des failles selon la nature du handicap et les particularités liées à l’état de santé, l’allégeance politique, la région et les expériences en ce qui a trait à la COVID‑19. Entre autres éléments, l’allégeance au Parti libéral et le statut de bénéficiaire de la PCU se signalent comme étant associés à des points de vue plus positifs. Un examen plus approfondi des réactions qualitatives montre que ces points de vue sont également liés à des perspectives divergentes entourant les prestations et les dépenses gouvernementales, les divisions partisanes et autres clivages sociaux et culturels.

In this paper, I argue that a fundamental tradeoff between freedom and equality in human social organization is, in fact, a learnable schedule for the management of uncertainty for an individual agent operating in a social world pervaded with randomness. This implies that political organization and state structures may be viewed as different ways of handling uncertainty, each trading off freedom and equality in a slightly different manner, but each with an equivalent, and theoretically symmetric, balance between the two. I will argue that uncertainty in an intelligent agent arises from one of three primary sources. First, the agent’s logical, objective, mental model of the environment may be noisy. Second, subjective, affective appraisals, of other individuals, their behaviours, and their intentions may present substantial randomness. Finally, the relationship between the objective, real, external world, to the subjective, internal world of the agent may be blurry. As I will show, in any basic two-level Bayesian model of reasoning, these same three sources of uncertainty naturally arise as three learnable parameters. These three parameters govern the operation of the model, but must be traded-off against each other. Furthermore, I will give these three dimensions of uncertainty some theoretical validity, by showing how they pervasively occur across a wide swatch of literatures as three dimensions of political freedom and equality. The relationship between the parameters and the freedoms is a key step in gaining a better understanding of political movements. For example, I discuss how the transition from mechanical solidarity to organic solidarity during the industrial revolution in the 19th century, as described by Durkheim, is in fact a shift in the management of uncertainty to one focused on immediate meanings to one focused on trust. Further, such shifts may be purposefully initiated by politically interested groups, by spreading new narratives or introducing new policy. Viewed as uncertainty management, any shifts so induced should be carefully evaluated in terms of what freedoms it compromises, with smoother transitions assured for shifts that remain in a position of equilibrium with respect to freedom and equality.

A crucial step in the analysis of persistent homology is the transformation of data into an appropriate topological object (in our case, a simplicial complex). Modern packages for persistent homology often construct Vietoris--Rips or other distance-based simplicial complexes on point clouds because they are relatively easy to compute. We investigate alternative methods of constructing these complexes and the effects of making associated choices during simplicial-complex construction on the output of persistent-homology algorithms. We present two new methods for constructing simplicial complexes from two-dimensional geospatial data (such as maps). We apply these methods to a California precinct-level voting data set, demonstrating that our new constructions can capture geometric characteristics that are missed by distance-based constructions. Our new constructions can thus yield more interpretable persistence modules and barcodes for geospatial data. In particular, they are able to distinguish short-persistence features that occur only for a narrow range of distance scales (e.g., voting behaviors in densely populated cities) from short-persistence noise by incorporating information about other spatial relationships between precincts., Comment: 30 pages

Kiley and Vaisey (2020) recently published a method for assessing whether survey respondents appear to be changing their beliefs between waves or whether they instead appear to be repeating fixed responses with temporary local influences. This question is important because these processes reflect very different theoretical models of the evolution of “personal culture.” That is, if cultural beliefs are primarily public and responsive to external discourse, we should observe more updating as people respond to changes in their local environment. On the other hand, if cultural beliefs are primarily something learned early, then “settled dispositions” should be relatively resilient to change. In this paper, we build on Kiley and Vaisey (2020) and introduce an alternative method for distinguishing between cases where respondents appear be actively updating their responses and situations where respondents’ responses appear to be settled. This method, based on structural equation modeling, provides a close fit to the theoretical models outlined in Kiley and Vaisey (2020) and provides even stronger support for their claim that most survey responses reflect settled dispositions developed prior to adulthood.

Conjoint experiments aiming to estimate average marginal component effects and related quantities have become a standard tool for social scientists. However, existing solutions for power analyses to find appropriate sample sizes for such studies have various shortcomings and accordingly, explicit sample size planning is rare. Based on recent advances in statistical inference for factorial experiments, we derive simple yet generally applicable formulae to calculate power and minimum required sample sizes for testing average marginal component effects (AMCEs), conditional AMCEs, as well as interaction effects in forced-choice conjoint experiments. The only input needed are expected effect sizes. Our approach only assumes random sampling of individuals or randomization of profiles and avoids any parametric assumption. Furthermore, we show that clustering standard errors on individuals is not necessary and does not affect power. Our results caution against designing conjoint experiments with small sample sizes, especially for detecting heterogeneity and interactions. We provide an R package that implements our approach.

Social conformity can spread social norms and behaviors through a society. This research examines such a process geographically for conformity with the norm that citizens should vote and consequent voter turnout. A mathematical model for this process is developed based on the Laplace equation, and predictions are tested with qualitative and quantitative spatial analyses of state-level voter turnout in American presidential elections. Results show that the diffusion of conformist behavior affects the local degree of turnout and produces highly specific and predictable voting behavior patterns across the United States, confirming the model.

Automated image analysis has received increasing attention in social scientific research, yet existing scholarship has mostly covered the application of supervised learning to classify images into predefined categories. This study focuses on the task of unsupervised image clustering, which aims to automatically discover categories from unlabelled image data. We first review the steps to perform image clustering and then focus on one key challenge in this task—finding intermediate representations of images. We present several methods of extracting intermediate image representations, including the bag-of-visual-words model, self-supervised learning, and transfer learning (in particular, feature extraction with pretrained models). We compare these methods using various visual datasets, including images related to protests in China from Weibo, images about climate change on Instagram, and profile images of the Russian Internet Research Agency on Twitter. In addition, we propose a systematic way to interpret and validate clustering solutions. Results show that transfer learning significantly outperforms the other methods. The dataset used in the pretrained model critically determines what categories the algorithms can discover.

In this paper we illustrate the use of Data Science techniques to analyse complex human communication. In particular, we consider tweets from leaders of political parties as a dynamical proxy to political programmes and ideas. We also study the temporal evolution of their contents as a reaction to specific events. We analyse levels of positive and negative sentiment in the tweets using new tools adapted to social media. We also train an Artificial Intelligence to recognise the political affiliation of a tweet. The AI is able to predict the origin of the tweet with a precision in the range of 71-75%, and the political leaning (left or right) with a precision of around 90%. This study is meant to be viewed as a proof-of-concept of interdisciplinary nature, at the interface between Data Science and political analysis.

Conjoint analysis is an experimental technique that has become quite popular to understand peoples decisions in multi-dimensional decision-making processes. Despite the importance of power analysis for experimental techniques, current literature has largely disregarded statistical power considerations when designing conjoint experiments. The main goal of this article is to provide researchers and practitioners with a practical tool to calculate the statistical power of conjoint experiments. To this end, we first conducted an extensive literature review to understand how conjoint experiments are designed and gauge the plausible effect sizes discovered in the literature. Second, we formulate a data generating model that is sufficiently flexible to accommodate a wide range of conjoint designs and hypothesized effects. Third, we present the results of an extensive series of simulation experiments based on the previously formulated data generation process. Our results show that---even with relatively large sample size and the number of trials---conjoint experiments are not suited to draw inferences for experiments with large numbers of experimental conditions and relatively small effect sizes. Specifically, Type S and Type M errors are especially pronounced for experimental designs with relatively small effective sample sizes (< 3000) or a high number of levels (> 15) that find small but statistically significant effects (< 0.03). The proposed online tool based on the simulation results can be used by researchers to perform power analysis of their designs and hence achieve adequate design for future conjoint experiments.

Forecasts of US presidential elections have gained considerable attention in recent years. However, as became evident in 2016 with the victory of Donald Trump, most of them consider presidential elections only at the national level, neglecting that these are ultimately decided by the Electoral College. In order to improve accuracy, we believe that forecasts should instead address outcomes at the state-level to determine the eventual Electoral College winner. We develop a political economy model of the incumbent vote share across states based on different short- and long-term predictors, referring up to the end of the second quarter of election years. Testing it against election outcomes since 1980, our model correctly predicts the eventual election winner in 9 out of 10 cases – including 2016 –, with the 2000 election being the exception. For the 2020 election, it expects Trump to lose the Electoral College, as only 6.2 percent of simulated outcomes cross the required threshold of 270 Electoral Votes, with a mean prediction of 106 Electoral Votes.

The Bread and Peace model of US presidential elections estimates `VOTE`, the incumbent’s party’s share of the major-party vote, as a function of rates of personal-income growth and military fatalities during the incumbent president’s term. I replicate past work fitting the model to elections since 1952, confirming that income growth and war deaths account for most variation in VOTE. Using data available as of the end of October, I then develop 5 income-growth forecasts for Donald Trump’s presidential term and use them to forecast Trump’s share of the two-party vote in the 2020 election. My 5 forecasts range from 49.3% to 56.1% with a mean of 53.0%, notably higher than final polling averages of 45.6%–46.2% and a preliminary election result of 48%. An idiosyncratic factor such as the sitting president’s impeachment might account for the model overrating Donald Trump’s popular-vote performance.

This study uses a natural experiment and machine learning to examine how threats shift immigrants' political attention from their home country to host society. We trace how the September 11 attacks, an exogenous shock, increased Arab and Indian Americans' interest in U.S. domestic politics. We classified 5,684 Arab American and Indian American newspaper articles using machine learning. We estimated that three more articles on U.S. domestic politics were published daily in the post-9/11 period than in previous years. We found that the intervention produced a 46% smaller effect among the host population by analyzing 23,086 New York Times articles. The findings demonstrate how threats shift the locus of targeted immigrant groups' political attention.